From: Linus Walleij Date: Sat, 16 Apr 2022 22:45:49 +0000 (+0200) Subject: mmc: mmci: Break IRQ status loop when all zero X-Git-Tag: for-5.19/block-exec-2022-06-02~78^2~29 X-Git-Url: https://git.kernel.dk/?a=commitdiff_plain;h=f7ad75041ba9b36d8141a1c0dd527154ed10b96e;p=linux-block.git mmc: mmci: Break IRQ status loop when all zero We iterate an extra time through the IRQ status handling loop despite nothing had fired. Enabling the debug prints: mmci-pl18x 80005000.mmc: op 01 arg 00000000 flags 000000e1 mmci-pl18x 80005000.mmc: irq0 (data+cmd) 00000001 mmci-pl18x 80005000.mmc: irq0 (data+cmd) 00000000 mmci-pl18x 80005000.mmc: op 01 arg 40ff8080 flags 000000e1 mmci-pl18x 80005000.mmc: irq0 (data+cmd) 00000001 mmci-pl18x 80005000.mmc: irq0 (data+cmd) 00000000 It is pointless to loop through the function when status is zero. Just break the loop if the status is zero. Signed-off-by: Linus Walleij Link: https://lore.kernel.org/r/20220416224549.627623-1-linus.walleij@linaro.org Signed-off-by: Ulf Hansson --- diff --git a/drivers/mmc/host/mmci.c b/drivers/mmc/host/mmci.c index 45b8608c935c..f3cf3152a397 100644 --- a/drivers/mmc/host/mmci.c +++ b/drivers/mmc/host/mmci.c @@ -1619,6 +1619,8 @@ static irqreturn_t mmci_irq(int irq, void *dev_id) do { status = readl(host->base + MMCISTATUS); + if (!status) + break; if (host->singleirq) { if (status & host->mask1_reg)